A microbial diversity perspective within the Restoring Resilient Ecosystems project

Dig Deeper - Microbial Community Analysis

Introduction

Biodiversity loss, ecosystem degradation, and habitat destruction are increasingly linked to human-driven changes in land use, including urbanisation, agriculture, and the exploitation of natural resources (European Parliament, 2025; Jaureguiberry et al., 2022). In response, governments across Europe — including the EU — have introduced ambitious environmental strategies such as the EU Biodiversity Strategy for 2030 (European Parliament, 2025) and the 30x30 target (Markwick, 2023), which aims to protect 30% of land and sea by the year 2030.

Ecological restoration plays a vital role in addressing these challenges. Rather than simply returning ecosystems to a previous state, modern approaches focus on restoring ecological processes and enhancing ecosystem resilience (Hicks, 2023).

The RestREco Initiative

RestREco (Restoring Resilient Ecosystems) is a NERC-funded research project that adopts a resilience-based perspective on ecological restoration. The initiative brings together researchers from:

• Cranfield University
  
• University of Stirling
• UK Centre for Ecology & Hydrology
  
• The National Trust
  
• Forest Research

Using a natural experiment design, RestREco studies a network of 133 ecological restoration sites across England and Scotland. The project aims to identify key drivers of ecosystem development, such as:

• Time since restoration began
  
• Initial ecological conditions
  
• Proximity to existing woodland and grassland

The goal is to understand how these factors influence ecosystem complexity, function, and resilience to future pressures (RestREco, 2024).

The Dig Deeper Study

As part of the RestREco initiative, the Dig Deeper study focused on how the age of restoration, establishment type, and site management affect soil microbial communities, specifically bacteria and fungi.

To explore this, high-throughput sequencing was conducted on:

• 16S rRNA gene (for bacterial communities)
  
• ITS region (for fungal communities)

The analysis focused on three main aspects:

• Alpha and beta diversity
  
• Taxonomic composition
  
• Functional diversity

These microbial assessments complement broader ecosystem-level measurements within the RestREco project, including vegetation, invertebrates, and ecosystem functions such as litter decomposition, pollination services, and soil thermodynamic efficiency.

The following sections describe the sampling design, metadata structure, and the processing pipeline used to characterise microbial communities.

Research Questions and Hypotheses

This study investigates how grassland site age, establishment method, management practices, and soil pH influence the diversity, taxonomic composition, and functional profiles of bacterial and fungal communities during restoration. It also explores the interactions between bacteria and fungi, focusing on potential correlations between microbial taxa, functional pathways, and fungal guilds.

The following hypotheses were formulated:

1. Site age, establishment method (green hay/bush (GH), seed mix (SM), natural regeneration (NR)), management type (cutting, ploughing, sheep, cattle), and pH affect alpha and beta diversity in bacterial and fungal communities.
2. These same variables influence the taxonomic composition of soil microbial communities.
3. Age, establishment method, and pH influence the functional diversity of bacterial and fungal communities.
4. There are correlations in taxon abundance between bacterial and fungal groups.
5. There are correlations in the abundance of functional pathways and guilds between bacterial and fungal communities.

---

Sample Design and Metadata Overview

Sample Collection and Geographic Coverage

A total of 330 soil samples were collected in 66 sites of England for each marker (5 per site). Although some were removed due to being incomplete (missing from the metadata).

Sample Zone - Based on GPS coordinates

Overview of the soil sampling and sequencing strategy for each microbial marker.

Metric	16S	ITS
Microbial group	Bacteria	Fungi
Region sampled	England	England
Number of sites	66	66
Samples per site	5	5
Total samples	330	330
Average reads per sample	~65,000	~65,000
Read count range	30,000–85,000	10,000–90,000

Sampling Summary

Metadata Overview

Each sample collected was accompanied by metadata capturing key environmental and management variables. These contextual factors were essential for interpreting variation in microbial diversity.

Description of metadata variables associated with soil samples

Variable	Description
Site	Name of the sampling site
Plot number	Subdivision of each site (usually 5 plots per site)
CU Code	Unique code for each sample
Year_est	Year of establishment of the site
Age	Site age (ranging from 1 year to over 100 years )
Latitude/Longitude	GPS coordinates of the sample
Establishment	Restoration type or land management
pH	Soil pH value at the time of sampling
EC	Electrical conductivity of the soil
Cutting	Whether the site is cut (1 = Yes, 0 = No)
Cattle	Presence of cattle grazing (1 = Yes, 0 = No)
Sheep	Presence of sheep grazing (1 = Yes, 0 = No)
Plough	Whether the soil has been ploughed (1 = Yes, 0 = No)

Metadata Summary

---

Methods

Source data

A total of 330 soil samples across 66 sites (five samples per site) were collected by the RestREco project. Soil was sieved to 2 mm and then frozen before DNA extraction. DNA was extracted using a QIAGEN PowerSoil kit. The V4–V5 region of the bacterial 16S gene and the ITS1–1F region of the fungal ITS gene were sequenced by Novagene using Illumina paired-end reads (250 bp) on Illumina NovaSeq 6000. The primer sequences used were:

• GTGCCAGCMGCCGCGGTAA / CCGTCAATTCCTTTGAGTTT

A total of 330 paired FASTQ files were provided for each amplicon. Metadata files included Ordinance Survey location, latitude and longitude, year of establishment, age, establishment method, chemical and physical soil properties, and management method (cutting, cattle, sheep, plough).

16S

Initial quality control was performed using FastQC v0.12.1 (Andrews, 2010) and summarised with MultiQC v1.14 (Ewels et al., 2016). A custom Bash script created a QIIME2-compatible manifest (see S02_qc.sh). Some samples not in metadata were removed.

Raw sequences were imported into QIIME2 v2022.10 (Bolyen et al., 2019). Primer/barcode removal was done using CutAdapt v4.4 (Martin, 2011). Quality profiles were visualised with qiime demux summarize.

Denoising and merging were done using DADA2 (QIIME2 v2024.2) (Callahan et al., 2016). ASVs with <10 reads and singletons were filtered. Sequences were aligned using MAFFT v7.505 (Katoh & Standley, 2013), and trees were built with FastTree v2.1.11 (Price et al., 2010).

Feature tables were grouped by site using median-ceiling values. Rarefaction was applied at 9473 reads/sample. Diversity metrics calculated included:

• Alpha: Observed Features, Shannon, Evenness, Faith’s PD
• Beta: Bray-Curtis, Jaccard, Unweighted/Weighted UniFrac

Statistical tests included Kruskal-Wallis for alpha, PERMANOVA for beta.

Taxonomy was assigned using a Naive Bayes classifier trained on Greengenes 13_8 (DeSantis et al., 2006). Differential abundance was analysed via ANCOM, collapsed to genus and family levels.

Functional predictions were generated using PICRUSt2 (QIIME2 v2024.5) (Douglas et al., 2020), followed by relative abundance, heatmaps, and differential testing. Alpha and beta functional diversity metrics were also calculated.

ITS

ITS reads were processed with ITSxpress v2.1.4 (Einarsson & Rivers, 2024), quality-checked via FastQC and MultiQC. Trimmed reads were imported into QIIME2 v2024.10, denoised with DADA2, and classified using a Naive Bayes classifier trained on UNITE v10 (Abarenkov et al., 2025).

Diversity metrics were computed at 30,000 reads/sample, visualised via Emperor plots, and tested using Kruskal-Wallis (alpha) and PERMANOVA (beta).

Taxa tables were collapsed to genus level and analysed with ANCOM. Fungal ecological roles were assigned using FUNGuild v1.1 (Nguyen et al., 2016), with classifications handled by a custom Python script (FUN_Search.py).

Functional diversity (alpha and beta) based on fungal guilds was assessed at a rarefaction depth of 11,000 reads/sample, with statistics via Kruskal-Wallis and PERMANOVA.

16S–ITS correlations

Correlations were computed in R v4.4.2 using psych::corr.test (Revelle, 2025). Data came from genus-level ASV feature tables and functional assignments (PICRUSt2 and FUNGuild).

Correlations were filtered based on:

• Strength: \(r > 0.7\) or \(r < -0.4\)
• Significance: adjusted \(p < 1 \times 10^{-20}\)

Low-abundance taxa (< median counts) were excluded for robustness. For functional correlations, thresholds of:

• Bacterial pathways: mean > 269,508
• Fungal guilds: mean > 3946

were applied.

Final analyses were performed on both strict and extended filtered datasets.

Scripts are available at: GitHub – dig-deeper-project

Summary of Pipeline

Workflow (16S)

Workflow (ITS)

---

Data Pre-Processing

MutliQC on raw data

Bacteria (16S)

You can explore the full MultiQC report by clicking the image below:

MultiQC Plot (16S)

’)

Fungi (ITS)

You can explore the full MultiQC report by clicking the image below:

MultiQC Plot (ITS)

’)

QC after denoising

Statistics Table Summary

After denoising, quality control (QC) metrics provide essential insights into the effectiveness of the data processing steps and the overall quality of the resulting feature table. This summary table presents key statistics for each sample, including the number of input reads, filtered reads, and final feature counts after denoising. These metrics help assess sequencing success, identify potential outliers, and ensure that sufficient data remain for robust downstream analyses. Samples with unusually low read counts or feature richness may need to be excluded or interpreted with caution.

Bacteria (16S)

Here is a link to the statistics after denoising to view it on QIIME2 (16S) : Statitics after denoising (16S)

Statitics after denoising (16S)

Fungi (ITS)

Here is a link to the statistics after denoising to view it on QIIME2 (ITS) : Statitics after denoising (16S)

Statitics after denoising (ITS)

QC Plots

These Quality Control (QC) plots were generated after trimming the sequencing reads. They provide a visual summary of the base quality scores, read length distributions, and other metrics, helping to assess whether the trimming step successfully removed low-quality regions and adapter contamination. Consistently high-quality reads across samples are essential for reliable downstream analysis.

Bacteria (16S)

QC plot (16S)

Fungi (ITS)

QC plot (ITS)

Rarefication Curves

Rarefaction curves provide a visual tool to assess sequencing depth and compare species richness between samples. In this study, rarefaction curves were generated separately for bacterial (16S rRNA gene) and fungal (ITS) communities, using the number of observed features—i.e., unique ASVs—as a proxy for richness.

For both bacteria and fungi, the shape of each curve indicates whether sequencing depth was sufficient to capture most of the diversity in a given sample. Curves that level off suggest that a representative portion of the community has been sampled, whereas rising curves indicate that additional sequencing could reveal further diversity.

This step is crucial to ensure that downstream diversity analyses are not biased by unequal sampling effort.

Bacteria (16S)

Here is a link to the Rarefiction plots for more flexibility on QIIME2: Rarefiction plots (16S)

Select variable: Establishment Age Year of Establishment Os Location Plough Lat/long

Rarefaction Curves of Observed Features by 'Establishment' (16S)

Rarefaction Curves of Observed Features by 'Age' (16S)

Rarefaction Curves of Observed Features by 'Year_est' (16S)

Rarefaction Curves of Observed Features by 'OS_location' (16S)

Rarefaction Curves of Observed Features by 'Lat_long' (16S)

Rarefaction Curves of Observed Features by 'Plough' (16S)

---

Fungi (ITS)

Here is a link to the Rarefiction plots for more flexibility on QIIME2: Rarefiction plots (ITS)

Select variable: Os Location CU Code Lat/long

Rarefaction Curves of Observed Features by 'OS_location' (ITS)

Rarefaction Curves of Observed Features by 'CU Code' (ITS)

Rarefaction Curves of Observed Features by 'Lat_long' (ITS)

---

Throughout the report, the median pH value used to define both pH_binary and pH_category is 7.94, while the threshold for both Age_binary and Age_category is 13 years.

Taxonomic Diversity (Bacteria & Fungi)

Results Summary
Microbial taxonomic diversity revealed clear responses to restoration strategies. For bacteria, establishment method was the strongest driver of alpha and beta diversity, followed by pH and grazing. For fungi, pH and age were more influential. NR sites generally showed lower bacterial diversity and functional richness than GH and SM.

Alpha Diversity

Alpha diversity refers to the variety of organisms within a particular sample or environment. It reflects both richness—the number of distinct taxa—and evenness—how evenly individuals are distributed among those taxa.

To allow interactive exploration of alpha diversity metrics across different environmental variables, we implemented a drop-down menu that dynamically displays the corresponding plots. Some variables, such as pH category, are only present in the ITS dataset, while others, like Year group, are specific to the 16S dataset. Internally, variables are mapped to their dataset-specific equivalents where needed (e.g. Age group in 16S becomes Age category in ITS). It is important to note, however, that these variables are not always directly comparable: for instance, Age group (16S) divides sites into multiple discrete intervals based on restoration age, while Age category (ITS) is a binary classification based on whether a site is above or below the median age ( Age_binary for bacteria). Despite these differences, the interface ensures that only available and relevant plots are shown for each selection.

For the bacteria plots, the data was aggragated by sites, while they were not for the fungi.

Bacteria (16S)

Results Summary

Within group diversity was assessed using Faith’s Phylogenetic Diversity (PD), Shannon entropy, and Pielou’s evenness. Significant differences were observed between establishment methods across all three metrics, while pH, management type, and site age showed no significant effects.

Shannon entropy differed significantly among establishment methods (p = 0.001), with green hay/bush (GH) sites showing higher diversity than natural regeneration (NR) (p-adjusted = 0.0019) and seed mix (SM) sites (p-adjusted = 0.036). SM also had higher diversity than NR (q = 0.0199).

Pielou’s evenness was also affected by establishment method (p = 0.0148). Pairwise differences were significant between GH and NR (p-adjusted = 0.0148), while the SM–NR comparison approached significance (p-adjusted = 0.0597). No significant difference was found between GH and SM

Faith’s PD also showed significant variation across establishment types (p = 0.0106), with GH sites exhibiting higher phylogenetic diversity than NR (q = 0.0148), and SM also higher than NR (q = 0.043). The difference between GH and SM was not significant.

Select variable: Establishment Cutting Cattle Sheep Plough Year group Age group pH binary Age binary

Shannon Index Boxplots

Click for more information on Shannon index

The Shannon index takes into account not only the number of species present, but also how evenly their abundances are distributed. A higher Shannon value generally indicates a more diverse and ecologically balanced community.

Here is a link to the full QIIME2 results (16S) : Shannon Index (16S)

Kruskal-Wallis p-value: 0.00102

Kruskal-Wallis p-value: 0.158

Kruskal-Wallis p-value: 0.939

Kruskal-Wallis p-value: 0.655

Kruskal-Wallis p-value: 0.316

Kruskal-Wallis p-value: 0.408

Kruskal-Wallis p-value: 0.126

Kruskal-Wallis p-value: 0.116

Kruskal-Wallis p-value: 0.358

Faith’s PD Boxplots

Click for more information on Faith’s PD

Faith’s Phylogenetic Diversity (Faith’s PD) measures the total branch length of the phylogenetic tree that spans the species in a sample. Unlike the Shannon index, Faith PD incorporates evolutionary relationships, providing a phylogenetic perspective on diversity. (Bacteria only)

Here is a link to the full QIIME2 results (16S) : Faith PD (16S)

Kruskal-Wallis p-value: 0.0106

Kruskal-Wallis p-value: 0.362

Kruskal-Wallis p-value: 0.878

Kruskal-Wallis p-value: 0.604

Kruskal-Wallis p-value: 0.705

Kruskal-Wallis p-value: 0.216

Kruskal-Wallis p-value: 0.0709

Kruskal-Wallis p-value: 0.0893

Kruskal-Wallis p-value: 0.393

Evenness Index Boxplots

Click for more information on Pielou’s Evenness Index

Pielou’s Evenness index specifically quantifies how equally individual organisms are distributed across taxa. While Shannon integrates both richness and evenness, this metric isolates the evenness component, providing a complementary view of diversity patterns.

Here is a link to the full QIIME2 results (16S) : Pielou Evenness (16S)

Kruskal-Wallis p-value: 0.0148

Kruskal-Wallis p-value: 0.17

Kruskal-Wallis p-value: 0.878

Kruskal-Wallis p-value: 0.291

Kruskal-Wallis p-value: 0.64

Kruskal-Wallis p-value: 0.854

Kruskal-Wallis p-value: 0.608

Kruskal-Wallis p-value: 0.542

Kruskal-Wallis p-value: 0.0817

---

Fungi (ITS)

Results Summary

Fungal alpha diversity was moderately influenced by environmental variables, particularly soil pH and site age.
Although median Shannon diversity appeared slightly higher in natural regeneration (NR) plots, no significant differences were observed between establishment methods (GH, SM, NR). Similarly, management practices such as sheep grazing, cattle grazing, ploughing, and cutting had no detectable effect on fungal Shannon diversity.

However, soil pH showed a significant influence: alkaline soils (pH > 7.9) had significantly lower Shannon diversity compared to neutral or slightly acidic soils (p < 0.02, Kruskal-Wallis).
Site age also impacted fungal diversity, with older sites (age > 13 years) exhibiting a lower median Shannon index (p < 0.02), suggesting a reduction in taxonomic richness or evenness over time.

Regarding community evenness, only sheep grazing had a statistically significant effect. Sites with sheep showed a slightly higher median Pielou’s Evenness index (p = 0.03). No significant differences were detected for other variables.

Select variable: Establishment Cutting Cattle Sheep Plough Age category pH category

Shannon Index Boxplots

Click for more information on Shannon index

The Shannon index takes into account not only the number of species present, but also how evenly their abundances are distributed. A higher Shannon value generally indicates a more diverse and ecologically balanced community.

Here is a link to the full QIIME2 results (ITS) : Shannon Index (ITS)

Kruskal-Wallis p-value: 0.63

Kruskal-Wallis p-value: 0.317

Kruskal-Wallis p-value: 0.881

Kruskal-Wallis p-value: 0.197

Kruskal-Wallis p-value: 0.133

Kruskal-Wallis p-value: 0.0175

Kruskal-Wallis p-value: 0.0194

Evenness Index Boxplots

Click for more information on Pielou’s Evenness Index

Pielou’s Evenness index specifically quantifies how equally individual organisms are distributed across taxa. While Shannon integrates both richness and evenness, this metric isolates the evenness component, providing a complementary view of diversity patterns.

Here is a link to the full QIIME2 results (ITS) : Pielou Evenness (ITS)

Kruskal-Wallis p-value: 0.305

Kruskal-Wallis p-value: 0.447

Kruskal-Wallis p-value: 0.91

Kruskal-Wallis p-value: 0.0337

Kruskal-Wallis p-value: 0.885

Kruskal-Wallis p-value: 0.203

Kruskal-Wallis p-value: 0.83

---

Comparative Microbial Community Composition (Beta Diversity)

To explore differences in microbial communities, we often rely on dimensionality reduction techniques such as Principal Coordinates Analysis (PCoA), visualised through Emperor plots. Two commonly used distance metrics in this context are Bray-Curtis and Jaccard.

While both metrics can reveal meaningful clustering and separation in microbial data, they capture complementary aspects of community structure.

This approach is useful to visualise group clustering by variables like establishment type, management, or site age, and complements statistical tests like PERMANOVA, which assess whether community structure significantly varies across those factors.

Results Summary

Beta diversity patterns confirmed that establishment type, pH, and grazing significantly shaped microbial community composition. For bacteria, strong clustering by method and significant PERMANOVA results support distinct assemblages. Fungal beta diversity followed similar trends but with additional sensitivity to site age.

---

Bray-Curtis

Click for more information on Bray-Curtis Distance Metric

Bray-Curtis is a non-phylogenetic beta diversity metric that compares microbial communities based on species abundance, taking into account both shared taxa and their relative proportions. Unlike presence/absence metrics, Bray-Curtis gives more weight to dominant species and is sensitive to changes in abundance structure. It ranges from 0 (identical composition and abundance) to 1 (completely different), reflecting how much two communities differ in the quantity of each taxon they contain. Bray-Curtis is especially useful for detecting shifts in community structure driven by changes in resource availability, disturbance, or other factors that influence the balance of taxa rather than just their presence.

Emperor Plot

Bacteria

Results Summary

Bray–Curtis PCoA plots revealed a strong separation of bacterial communities according to restoration establishment method, with NR, GH, and SM forming distinguishable clusters, although some overlap was present. Soil pH also showed a clear gradient in community composition, while site age produced only weak visual separation, with no consistent clustering of younger versus older sites. Sheep grazing did not produce clearly distinct clusters in ordination space, though minor shifts were observable.

Here is a link to the Bray-Curtis Emperor Plot for more flexibility on QIIME2: Bray-Curtis Emperor Plot (16S)

Bray-Curtis Emperor Plot

Fungi

Results Summary

Beta diversity analysis revealed that establishment method, soil pH, and site age had the most significant influence on fungal community composition, followed by sheep grazing, ploughing, and cutting. In PCoA plots, natural regeneration sites formed a loose cluster near seed mix sites, while green hay/bush sites clustered more distinctly along Axis 1.

Here is a link to the Bray-Curtis Emperor Plot for more flexibility on QIIME2: Bray-Curtis Emperor Plot (ITS)

Bray-Curtis Emperor Plot (ITS)

PERMANOVA

Bacteria

Results Summary

Bray-Curtis-based PERMANOVA revealed significant differences in bacterial community composition across establishment methods, with all pairwise comparisons between methods also significant. These results indicate that restoration strategy has a strong influence on both the composition and abundance structure of bacterial communities.

Site age also had a significant but weaker effect. However, clustering by age in Bray-Curtis PCoA plots was less pronounced, suggesting more subtle shifts in community structure over time.

Soil pH significantly shaped bacterial community composition, with Bray-Curtis PCoA plots showing clear clustering between high and low-to-neutral pH soils

You can click on the images below to access the full QIIME2 report.

Select variable: Establishment Cutting Cattle Sheep Plough Year group pH binary Age binary

View full QIIME2 results (Bray–Curtis – Establishment)

Figure: PERMANOVA Bray–Curtis for “Establishment” (16S)

View full QIIME2 results (Bray–Curtis – Cutting)

Figure: PERMANOVA Bray–Curtis for “Cutting” (16S)

View full QIIME2 results (Bray–Curtis – Cattle)

Figure: PERMANOVA Bray–Curtis for “Cattle” (16S)

View full QIIME2 results (Bray–Curtis – Sheep)

Figure: PERMANOVA Bray–Curtis for “Sheep” (16S)

View full QIIME2 results (Bray–Curtis – Plough)

Figure: PERMANOVA Bray–Curtis for “Plough” (16S)

View full QIIME2 results (Bray–Curtis – Year_group)

Figure: PERMANOVA Bray–Curtis for “Year_group” (16S)

View full QIIME2 results (Bray–Curtis – pH_binary)

Figure: PERMANOVA Bray–Curtis for “pH_binary” (16S)

View full QIIME2 results (Bray–Curtis – Age_binary)

Figure: PERMANOVA Bray–Curtis for “Age_binary” (16S)

Fungi (ITS)

Results Summary

Statistical analysis with PERMANOVA based on Bray–Curtis distances confirmed significant effects of multiple variables on fungal community structure:

• Establishment method: p = 0.002

  • GH vs. SM: p = 0.008
    
  • GH vs. NR: p = 0.011

  These results show that assisted restoration strategies (GH, SM) led to distinct community abundance patterns, while NR sites displayed the highest variation, with dissimilarity values ranging from 0.25 to 0.95.

• Soil pH: p = 0.001
  The effect of pH was significant, with alkaline soils supporting less diverse but more homogeneous fungal communities compared to neutral or acidic sites.

• Site age: p = 0.024
  Differences were observed between younger and older sites, consistent with gradual shifts in fungal community structure over time.

• Sheep grazing: p = 0.025
  Grazing pressure appeared to influence community composition, possibly through soil disturbance or altered plant-fungal interactions.

Overall, Bray–Curtis-based analyses highlight how restoration strategy, soil conditions, and management practices interact to shape fungal abundance profiles across the landscape.

You can click on the images below to access the full QIIME2 report.

Select variable: Establishment Cutting Cattle Sheep Plough pH category Age category

View full QIIME2 results (Bray–Curtis – Establishment)

Figure: PERMANOVA Bray–Curtis for “Establishment” (ITS)

View full QIIME2 results (Bray–Curtis – Cutting)

Figure: PERMANOVA Bray–Curtis for “Cutting” (ITS)

View full QIIME2 results (Bray–Curtis – Cattle)

Figure: PERMANOVA Bray–Curtis for “Cattle” (ITS)

View full QIIME2 results (Bray–Curtis – Sheep)

Figure: PERMANOVA Bray–Curtis for “Sheep” (ITS)

View full QIIME2 results (Bray–Curtis – Plough)

Figure: PERMANOVA Bray–Curtis for “Plough” (ITS)

View full QIIME2 results (Bray–Curtis – pH_category)

Figure: PERMANOVA Bray–Curtis for “pH_category” (ITS)

View full QIIME2 results (Bray–Curtis – Age_category)

Figure: PERMANOVA Bray–Curtis for “Age_category” (ITS)

Jaccard

Click for more information on Jaccard Distance Metric

Jaccard is a non-phylogenetic beta diversity metric that compares microbial communities based on the presence or absence of taxa, without considering their abundance or evolutionary relationships. It measures the proportion of shared taxa between samples relative to the total number of taxa present, treating all species equally. This makes it particularly sensitive to community turnover, highlighting whether two sites contain the same organisms, regardless of how abundant they are. Jaccard is well-suited for detecting changes in community membership across environmental gradients or treatment groups, especially when interested in whether taxa are present at all, rather than how dominant they are.

Emperor Plot

Bacteria

Here is a link to the Jaccard Emperor Plot for more flexibility on QIIME2: Jaccard Emperor Plot (16S)

Jaccard Emperor Plot (16S)

Fungi

Results Summary

The Jaccard diversity index revealed a wider distribution of points across PCoA space for all tested variables, indicating high variability in community membership (presence/absence-based).

• Establishment method: Green hay/bush sites formed a loose cluster, with seed mix and natural regeneration showing partial co-clustering similar to the Bray–Curtis results. Natural regeneration sites exhibited stronger grouping in PCs 3 to 5, suggesting more consistent community membership among those sites in deeper dimensions.

• Soil pH: Sites appeared to form three clusters—one containing mostly alkaline soils, one with mixed pH profiles, and a third comprising three strongly acidic sites (pH 6.6–6.62), which separated distinctly.

• Site age: Two main clusters emerged—one made up mostly of younger sites (age < 13), and a second with mixed-age sites, suggesting some temporal structuring.

Here is a link to the Jaccard Emperor Plot for more flexibility on QIIME2: Jaccard Emperor Plot (ITS)

Jaccard Emperor Plot (ITS)

PERMANOVA

Bacteria (16S)

You can click on the images below to access the full QIIME2 report.

Select variable: Establishment Cutting Cattle Sheep Plough Year_group pH binary Age binary

View full QIIME2 results (Jaccard – Establishment)

Figure: PERMANOVA Jaccard for “Establishment” (16S)

View full QIIME2 results (Jaccard – Cutting)

Figure: PERMANOVA Jaccard for “Cutting” (16S)

View full QIIME2 results (Jaccard – Cattle)

Figure: PERMANOVA Jaccard for “Cattle” (16S)

View full QIIME2 results (Jaccard – Sheep)

Figure: PERMANOVA Jaccard for “Sheep” (16S)

View full QIIME2 results (Jaccard – Plough)

Figure: PERMANOVA Jaccard for “Plough” (16S)

View full QIIME2 results (Jaccard – Year_group)

Figure: PERMANOVA Jaccard for “Year_group” (16S)

View full QIIME2 results (Jaccard – pH_binary)

Figure: PERMANOVA Jaccard for “pH_binary” (16S)

View full QIIME2 results (Jaccard – Age_binary)

Figure: PERMANOVA Jaccard for “Age_binary” (16S)

Fungi (ITS)

Results Summary

PERMANOVA analysis confirmed that fungal community composition was significantly influenced by multiple environmental factors:

• Establishment method had a strong effect (Bray–Curtis: p = 0.002; Jaccard: p = 0.001). Pairwise comparisons showed the most significant differences for:

  • Green hay/bush vs. seed mix (p = 0.008)
  • Green hay/bush vs. natural regeneration (p = 0.011)
  • NR vs. SM (p = 0.004, Jaccard)

  Notably, natural regeneration sites exhibited the greatest variability in fungal composition, with Bray–Curtis distances ranging from 0.25 to 0.95, and Jaccard medians exceeding 0.75 across establishment types.

• Soil pH was also a strong driver (Bray–Curtis: p = 0.001; Jaccard: p = 0.002), showing clear differences in fungal community structure between alkaline and less alkaline sites.

• Site age significantly influenced fungal beta diversity (Bray–Curtis: p = 0.024; Jaccard: p = 0.006), supporting the idea of successional community turnover.

• Management practices also had an effect:

  • Sheep grazing: p = 0.025 (Bray–Curtis)
  • Ploughing: p = 0.024 (Jaccard)
  • Cutting: p = 0.04 (Jaccard)

These findings highlight the combined influence of land-use strategy and edaphic conditions on fungal community turnover, particularly at the level of taxon presence/absence.

You can click on the images below to access the full QIIME2 report.

Select variable: Establishment Cutting Cattle Sheep Plough pH category Age category

View full QIIME2 results (Jaccard – Establishment)

Figure: PERMANOVA Jaccard for “Establishment” (ITS)

View full QIIME2 results (Jaccard – Cutting)

Figure: PERMANOVA Jaccard for “Cutting” (ITS)

View full QIIME2 results (Jaccard – Cattle)

Figure: PERMANOVA Jaccard for “Cattle” (ITS)

View full QIIME2 results (Jaccard – Sheep)

Figure: PERMANOVA Jaccard for “Sheep” (ITS)

View full QIIME2 results (Jaccard – Plough)

Figure: PERMANOVA Jaccard for “Plough” (ITS)

View full QIIME2 results (Jaccard – pH_category)

Figure: PERMANOVA Jaccard for “pH_category” (ITS)

View full QIIME2 results (Jaccard – Age_category)

Figure: PERMANOVA Jaccard for “Age_category” (ITS)

---

UniFrac metrics (both weighted and unweighted) rely on a phylogenetic tree to quantify the evolutionary distances between microbial taxa. Since the ITS region used for fungal community profiling does not provide reliable phylogenetic resolution, it is not suitable for generating accurate phylogenetic trees. As a result, UniFrac analyses were only conducted for 16S bacterial data, and not for fungi.

Unweighted Unifrac - Bacteria Only (16S)

Click for more information on Unweighted Unifrac Distance Metric

Unweighted UniFrac is a phylogenetic beta diversity metric that compares microbial communities based solely on the presence or absence of taxa, while incorporating their evolutionary relationships.

Unlike abundance-based measures, this metric considers whether lineages are shared between communities, regardless of how dominant they are. It is particularly sensitive to rare or low-abundance taxa, as all taxa are weighted equally.

Unweighted UniFrac is useful for identifying broad differences in community membership—for example, whether two sites share the same species, even if those species occur at very different abundances. This makes it well-suited to detecting compositional shifts due to strong environmental filters or historical legacies.

Results Summary

PCoA based on Unweighted UniFrac distances showed that bacterial communities clustered primarily by establishment method, with separation between NR, GH, and SM. A small subset of NR samples appeared distinct from the main NR cluster, suggesting within-group heterogeneity. In contrast, site age and sheep grazing showed little to no clear visual separation in the top three axes, and pH presented only a mild gradient.

Emperor Plot - Bacteria

Here is a link to the Unweighted Unifrac Emperor Plot for more flexibility on QIIME2: Unweighted Unifrac Emperor Plot (16S)

Unweighted Unifrac Emperor Plot (16S)

PERMANOVA - Bacteria

Unweighted UniFrac PERMANOVA showed that establishment method, soil pH, and sheep grazing all significantly influenced bacterial community membership (all p = 0.001 or 0.002), with strong effects across establishment types (q < 0.005 for all pairwise comparisons). In contrast, site age was not significant (p-adjusted = 0.063), suggesting that phylogenetic composition remains relatively stable across the age gradient, even if abundance shifts are occurring.

You can click on the images below to access the full QIIME2 report.

Select variable: Establishment Cutting Cattle Sheep Plough Year_group pH binary Age binary

View full QIIME2 results (Unweighted Unifrac – Establishment)

Figure: PERMANOVA Unweighted Unifrac for “Establishment” (16S)

View full QIIME2 results (Unweighted Unifrac – Cutting)

Figure: PERMANOVA Unweighted Unifrac for “Cutting” (16S)

View full QIIME2 results (Unweighted Unifrac – Cattle)

Figure: PERMANOVA Unweighted Unifrac for “Cattle” (16S)

View full QIIME2 results (Unweighted Unifrac – Sheep)

Figure: PERMANOVA Unweighted Unifrac for “Sheep” (16S)

View full QIIME2 results (Unweighted Unifrac – Plough)

Figure: PERMANOVA Unweighted Unifrac for “Plough” (16S)

View full QIIME2 results (Unweighted Unifrac – Year_group)

Figure: PERMANOVA Unweighted Unifrac for “Year_group” (16S)

View full QIIME2 results (Unweighted Unifrac – pH_binary)

Figure: PERMANOVA Unweighted Unifrac for “pH_binary” (16S)

View full QIIME2 results (Unweighted Unifrac – Age_binary)

Figure: PERMANOVA Unweighted Unifrac for “Age_binary” (16S)

Weighted Unifrac - Bacteria Only (16S)

Click for more information on Weighted Unifrac Distance Metric

Weighted UniFrac is a phylogenetic distance metric that takes into account both the evolutionary relationships between taxa and their relative abundances in each sample.

It quantifies how much of the phylogenetic tree is shared between communities, with branches weighted by the proportion of abundance they represent. As a result, changes in dominant taxa have more influence on the distance than rare species.

Weighted UniFrac is ideal for detecting differences in community structure, especially when those differences involve shifts in abundant lineages. It provides a more nuanced view than unweighted metrics by integrating both phylogenetic and quantitative information.

Emperor Plot - Bacteria

Here is a link to the Weighted Unifrac Emperor Plot for more flexibility on QIIME2: Weighted Unifrac Emperor Plot (16S)

Weighted Unifrac Emperor Plot (16S)

PERMANOVA - Bacteria

You can click on the images below to access the full QIIME2 report.

Select variable: Establishment Cutting Cattle Sheep Plough Year_group pH binary Age binary

View full QIIME2 results (Weighted Unifrac – Establishment)

Figure: PERMANOVA Weighted Unifrac for “Establishment” (16S)

View full QIIME2 results (Weighted Unifrac – Cutting)

Figure: PERMANOVA Weighted Unifrac for “Cutting” (16S)

View full QIIME2 results (Weighted Unifrac – Cattle)

Figure: PERMANOVA Weighted Unifrac for “Cattle” (16S)

View full QIIME2 results (Weighted Unifrac – Sheep)

Figure: PERMANOVA Weighted Unifrac for “Sheep” (16S)

View full QIIME2 results (Weighted Unifrac – Plough)

Figure: PERMANOVA Weighted Unifrac for “Plough” (16S)

View full QIIME2 results (Weighted Unifrac – Year_group)

Figure: PERMANOVA Weighted Unifrac for “Year_group” (16S)

View full QIIME2 results (Weighted Unifrac – pH_binary)

Figure: PERMANOVA Weighted Unifrac for “pH_binary” (16S)

View full QIIME2 results (Weighted Unifrac – Age_binary)

Figure: PERMANOVA Weighted Unifrac for “Age_binary” (16S)

---

Taxonomy Composition

Taxonomy Barplot

Bacteria (16S)

Results Summary

Analysis of bacterial community composition at the phylum level across establishment types revealed that Actinobacteria and Proteobacteria were dominant, together accounting for 50–80% of relative abundance. Actinobacteria were particularly abundant in seed mix (SM) sites, where they reached up to ~48%, while their proportion was lowest in natural regeneration (NR) sites (~21%). In contrast, Proteobacteria were most prominent in NR soils, peaking at ~55%. Other frequently detected phyla included Acidobacteria, Firmicutes, Chloroflexi, Verrucomicrobia, Planctomycetes, and Nitrospirae, with Verrucomicrobia notably more abundant in NR plots.

At the class level, Alphaproteobacteria, Actinobacteria, Thermoleophilia, and Bacilli dominated across sites. A higher proportion of Spartobacteria was observed in NR plots compared to GH and SM plots.

When comparing soils with pH above and below the median, the dominant phyla remained broadly consistent. However, some shifts were noted: Firmicutes and Planctomycetes were more abundant in more acidic soils (below-median pH), with Firmicutes reaching up to ~30% (versus ~19% above the median), and Planctomycetes peaking at ~6.5% (compared to ~3.6%).

Here is a link to the Taxonomy Barplots for more flexibility on QIIME2: Taxonomy Barplot (16S)

Select variable: Establishment Cutting Cattle Sheep Plough Age group pH binary Year group Age binary

Taxonomy Barplots Associated with 'Establishment' (16S)

Taxonomy Barplots Associated with 'Cutting' (16S)

Taxonomy Barplots Associated with 'Cattle' (16S)

Taxonomy Barplots Associated with 'Sheep' (16S)

Taxonomy Barplots Associated with 'Plough' (16S)

Taxonomy Barplots Associated with 'Age_group' (16S)

Taxonomy Barplots Associated with 'pH_binary' (16S)

Taxonomy Barplots Associated with 'Year_group' (16S)

Taxonomy Barplots Associated with 'Age_binary' (16S)

Fungi (ITS)

Results Summary

For most of the studied sites, the dominant fungal phyla were Ascomycota, Mortierellomycota, and Basidiomycota, while a large proportion of sequences were classified as incertae sedis, indicating uncertain phylogenetic placement. Some variation in taxonomic composition was observed in natural regeneration (NR) sites, where Basidiomycota appeared more prominently in at least one site.

At the class level, the most commonly detected groups included Sordariomycetes, Mortierellomycetes, Dothideomycetes, and Leotiomycetes. NR sites generally exhibited a lower frequency of Agaricomycetes, except for a single outlier, compared to sites restored using green hay (GH) or seed mix (SM).

When examining fungal composition by pH category, the only consistent pattern observed was a higher frequency of Sordariomycetes in slightly acidic to neutral soils (below median pH) compared to more alkaline soils.

Here is a link to the Taxonomy Barplots for more flexibility on QIIME2: Taxonomy Barplot (ITS)

Select variable: Establishment Cutting Cattle Sheep Plough Age category pH category

Taxonomy Barplots Associated with 'Establishment' (ITS)

Taxonomy Barplots Associated with 'Cutting' (ITS)

Taxonomy Barplots Associated with 'Cattle' (ITS)

Taxonomy Barplots Associated with 'Sheep' (ITS)

Taxonomy Barplots Associated with 'Plough' (ITS)

Taxonomy Barplots Associated with 'Age_category' (ITS)

Taxonomy Barplots Associated with 'pH_category' (ITS)

Krona Plots

To explore the composition of soil microbial communities, we used Krona plots — interactive, circular charts that display taxonomic abundances in a hierarchical manner.

These plots allow users to intuitively navigate from broader taxonomic levels (such as Phylum) to more specific ones (like Genus), while simultaneously comparing relative abundances across taxa.

In this study, Krona plots provide a powerful and user-friendly way to:

• Visualise which microbial groups dominate each site
• Explore the taxonomic diversity present in bacterial, archaeal, and fungal communities

You can click on the images below to access the Krona plots for each site.

Bacteria (16S)

Krona Plot for Baltic_farm_1 (16S)

Fungi (ITS)

Krona Plot for Baltic_farm_1 (ITS)

Differential Abundance Analysis with ANCOM

We used ANCOM to identify taxa whose relative abundances significantly differed across groups. This method accounts for the compositional nature of microbiome data by comparing log-ratios between taxa. The results are shown as volcano-like plots, where the W statistic reflects how many pairwise comparisons a taxon was found to differ in. Significant taxa are highlighted accordingly.

Bacteria (16S)

Results Summary

ANCOM analysis identified several bacterial taxa whose abundance varied significantly across environmental factors.

Conexibacteraceae was more abundant in soils with below-median pH, indicating a preference for more acidic conditions.

The genus Tetrasphaera showed significantly higher abundance in younger sites, suggesting a potential role in early stages of soil restoration.

Two taxa were strongly associated with establishment method: an unclassified genus within the Intrasporangiaceae family and the genus Blastococcus were both more abundant in SM and GH plots compared to NR. These differences were supported by large W statistics and differences in median abundance values across treatments. These taxa responded notably to restoration strategy, indicating that establishment method can drive specific shifts in bacterial community structure.

Select variable: Establishment Cutting Cattle Sheep Plough Age_group pH_binary Year_group Age_binary

View interactive volcano-like plot in QIIME2 (16S - Establishment )

Differences in Taxa Abundance Associated with 'Establishment' (ANCOM Results) (16S)

View interactive volcano-like plot in QIIME2 (16S - Cutting )

Differences in Taxa Abundance Associated with 'Cutting' (ANCOM Results) (16S)

View interactive volcano-like plot in QIIME2 (16S - Cattle )

Differences in Taxa Abundance Associated with 'Cattle' (ANCOM Results) (16S)

View interactive volcano-like plot in QIIME2 (16S - Sheep )

Differences in Taxa Abundance Associated with 'Sheep' (ANCOM Results) (16S)

View interactive volcano-like plot in QIIME2 (16S - Plough )

Differences in Taxa Abundance Associated with 'Plough' (ANCOM Results) (16S)

View interactive volcano-like plot in QIIME2 (16S - Age_group )

Differences in Taxa Abundance Associated with 'Age_group' (ANCOM Results) (16S)

View interactive volcano-like plot in QIIME2 (16S - pH_binary )

Differences in Taxa Abundance Associated with 'pH_binary' (ANCOM Results) (16S)

View interactive volcano-like plot in QIIME2 (16S - Year_group )

Differences in Taxa Abundance Associated with 'Year_group' (ANCOM Results) (16S)

View interactive volcano-like plot in QIIME2 (16S - Age_binary )

Differences in Taxa Abundance Associated with 'Age_binary' (ANCOM Results) (16S)

Fungi (ITS)

Results Summary

The fungal genus Metapochonia was identified as significantly differentially abundant between soils classified by pH category (above- vs. below-median). According to ANCOM results (W = 386), Metapochonia exhibited consistently higher abundance in soils with below-median pH (slightly acidic to neutral), with a notably higher median (33.0 vs. 8.0) and upper quartile abundance compared to soils with more alkaline pH.

ANCOM also revealed significant differences in fungal abundance between site age categories. A taxon from the Lasiosphaeriaceae family (W = 366) was more abundant in younger soils (below-median age), whereas the genus Paraphaeosphaeria (W = 320) was significantly enriched in older soils.

When comparing restoration establishment methods, two fungal taxa were found to differ significantly in abundance. The genus Gibellulopsis (W = 596) was most abundant in green hay (GH) sites (median = 200.5, max = 2118.0), followed by seed mix (SM) sites (median = 155.5), and was largely absent from natural regeneration (NR) sites. Conversely, Metarhizium (W = 535) showed its highest abundance in NR sites (median = 14.5, max = 327.0), with negligible presence in GH and SM sites.

Select variable: Establishment Cutting Cattle Sheep Plough Age_category pH_category

View interactive volcano-like plot in QIIME2 (ITS - Establishment )

Differences in Taxa Abundance Associated with 'Establishment' (ANCOM Results) (ITS)

View interactive volcano-like plot in QIIME2 (ITS - Cutting )

Differences in Taxa Abundance Associated with 'Cutting' (ANCOM Results) (ITS)

View interactive volcano-like plot in QIIME2 (ITS - Cattle )

Differences in Taxa Abundance Associated with 'Cattle' (ANCOM Results) (ITS)

View interactive volcano-like plot in QIIME2 (ITS - Sheep )

Differences in Taxa Abundance Associated with 'Sheep' (ANCOM Results) (ITS)

View interactive volcano-like plot in QIIME2 (ITS - Plough )

Differences in Taxa Abundance Associated with 'Plough' (ANCOM Results) (ITS)

View interactive volcano-like plot in QIIME2 (ITS - Age_category )

Differences in Taxa Abundance Associated with 'Age_category' (ANCOM Results) (ITS)

View interactive volcano-like plot in QIIME2 (ITS - pH_category )

Differences in Taxa Abundance Associated with 'pH_category' (ANCOM Results) (ITS)

---

Functional Diversity (Bacteria & Fungi)

Functional diversity refers to the variety of biological functions present in microbial communities. Unlike taxonomic analysis, which identifies who is there, functional analysis explores what they can do. This section includes two parallel analyses:

• Bacteria (16S): Functional profiles were inferred using PICRUSt2, predicting gene family and pathway abundance.
• Fungi (ITS): Functional roles were explored using FUNGuild, which assigns taxa to ecological guilds based on literature.

Alpha and beta diversity were analysed, along with differentially abundant functions or pathways.

Results Summary

Functional diversity revealed strong effects of establishment method and pH on microbial pathway composition. Core metabolic functions were consistent across sites, but specific biosynthetic and degradation pathways differed by age, pH, and grazing. Saprotrophic functions were dominant in fungi, with guild shifts linked to disturbance and soil chemistry.

Bacteria (16S)

Results Summary

Bacterial functional alpha and beta diversity varied significantly across establishment types, with GH and SM plots showing enhanced diversity. Functional profiles also responded to pH and age. Several metabolic pathways, including heme and amino acid biosynthesis, were enriched in specific conditions, reflecting adaptive responses to restoration.

Functional Alpha Diversity

We examined three complementary metrics: Shannon diversity (richness and evenness), Observed Features (richness only), and Evenness (dominance balance). You can use the menu below to explore how each metric varies according to different experimental variables.

Results Summary

Functional alpha diversity, measured by Shannon entropy of predicted metabolic pathways (PICRUSt2), differed significantly across establishment types (Kruskal-Wallis p = 0.0148). Pairwise tests showed that SM plots had significantly higher functional diversity than NR (p-adjusted = 0.0034), and GH was also higher than NR (p-adjusted = 0.02). No significant difference was observed between SM and GH (p-adjusted = 0.75).

Select variable: Establishment Cutting Cattle Sheep Plough Age_group pH_binary Age_binary

Shannon Index Boxplots

Here is a link to the Shannon Index Boxplot for more flexibility on QIIME2: Shannon Index Boxplot - Functional (16S)

Kruskal-Wallis p-value: 0.00164

Kruskal-Wallis p-value: 0.0496

Kruskal-Wallis p-value: 0.739

Kruskal-Wallis p-value: 0.681

Kruskal-Wallis p-value: 0.588

Kruskal-Wallis p-value: 0.144

Kruskal-Wallis p-value: 0.517

Kruskal-Wallis p-value: 0.0887

Observed Features Boxplots

Here is a link to the Observed features Boxplot for more flexibility on QIIME2: Observed Features Boxplot - Functional (16S)

Kruskal-Wallis p-value: 0.0354

Kruskal-Wallis p-value: 0.758

Kruskal-Wallis p-value: 0.969

Kruskal-Wallis p-value: 0.707

Kruskal-Wallis p-value: 0.923

Kruskal-Wallis p-value: 0.372

Kruskal-Wallis p-value: 0.292

Kruskal-Wallis p-value: 0.974

Evenness Index Boxplots

Here is a link to the Evenness Boxplot for more flexibility on QIIME2: Evenness Boxplot - Functional (16S)

Kruskal-Wallis p-value: 0.0294

Kruskal-Wallis p-value: 0.13

Kruskal-Wallis p-value: 0.888

Kruskal-Wallis p-value: 0.858

Kruskal-Wallis p-value: 0.393

Kruskal-Wallis p-value: 0.84

Kruskal-Wallis p-value: 0.184

Kruskal-Wallis p-value: 0.235

Functional Beta Diversity

Emperor Plot – Bray-Curtis Distance

Results Summary

Functional beta diversity analyses based on PICRUSt2 pathway profiles showed that multiple site-level factors significantly influenced microbial community composition. Establishment method had the strongest effect, with distinct functional profiles observed across all establishment types.

Beta diversity analysis of predicted microbial functions (Bray-Curtis) revealed that functional composition was significantly influenced by establishment method (p = 0.001), soil pH (p = 0.012), site age (p = 0.035), and grazing (p = 0.035). Pairwise comparisons confirmed significant differences between all establishment types (GH, SM, NR), while Emperor plots showed partial separation between high and low pH sites, and between grazed and ungrazed plots.

Here is a link to the Bray-Curtis Emperor Plot for more flexibility on QIIME2: Bray-Curtis Emperor Plot (16S)

Bray-Curtis Emperor Plot (Functional)

PERMANOVA – Bray-Curtis Distance

You can click on the images below to access the full QIIME2 report.

Select variable: Establishment Cutting Cattle Sheep Plough Year group Age group Age Binary pH Binary

View full QIIME2 results (Bray–Curtis – Establishment)

Figure: PERMANOVA Bray-Curtis for “Establishment” (ITS)

View full QIIME2 results (Bray–Curtis – Cutting)

Figure: PERMANOVA Bray-Curtis for “Cutting” (ITS)

View full QIIME2 results (Bray–Curtis – Cattle)

Figure: PERMANOVA Bray-Curtis for “Cattle” (ITS)

View full QIIME2 results (Bray–Curtis – Sheep)

Figure: PERMANOVA Bray-Curtis for “Sheep” (ITS)

View full QIIME2 results (Bray–Curtis – Plough)

Figure: PERMANOVA Bray-Curtis for “Plough” (ITS)

View full QIIME2 results (Bray–Curtis – Year_group)

Figure: PERMANOVA Bray-Curtis for “Year_group” (ITS)

View full QIIME2 results (Bray–Curtis – Age_group)

Figure: PERMANOVA Bray-Curtis for “Age_group” (ITS)

View full QIIME2 results (Bray–Curtis – Age_binary)

Figure: PERMANOVA Bray-Curtis for “Age_binary” (ITS)

View full QIIME2 results (Bray–Curtis – pH_binary)

Figure: PERMANOVA Bray-Curtis for “pH_binary” (ITS)

Differentially Abundant Pathways

To identify pathways with significantly different abundances between groups, ANCOM was applied to PICRUSt2 pathway predictions.

Results Summary

Differential abundance analysis of predicted microbial pathways revealed distinct functional shifts across environmental and management gradients.

Establishment method had a notable effect on microbial functional profiles, with two pathways significantly differing between strategies: PWY-5265 (peptidoglycan biosynthesis II in staphylococci) and PWY-6892 (thiazole component of thiamine diphosphate biosynthesis I).

Site age was associated with changes in core biosynthetic functions, including HEME-BIOSYNTHESIS-II, PWY-5918 (heme b biosynthesis), and HOMOSER-METSYN-PWY (L-methionine biosynthesis I).

Soil pH influenced two secondary metabolism pathways: PWY-6915 and PWY-6919, both involved in pentalenolactone biosynthesis.

Sheep grazing showed the strongest functional signal, with five pathways significantly enriched or depleted, notably: PWY-1541, P162-PWY, PWY-6944, PWY-3661, and PWY-5177, reflecting enhanced nitrogen and steroid metabolism as well as osmolyte degradation.

Select variable: Establishment Cutting Cattle Sheep Plough Age_group pH_binary Age_binary

Differences in Pathway Abundance Associated with 'Establishment' (ANCOM Results) (ITS)

Differences in Pathway Abundance Associated with 'Cutting' (ANCOM Results) (ITS)

Differences in Pathway Abundance Associated with 'Cattle' (ANCOM Results) (ITS)

Differences in Pathway Abundance Associated with 'Sheep' (ANCOM Results) (ITS)

Differences in Pathway Abundance Associated with 'Plough' (ANCOM Results) (ITS)

Differences in Pathway Abundance Associated with 'Age_group' (ANCOM Results) (ITS)

Differences in Pathway Abundance Associated with 'pH_binary' (ANCOM Results) (ITS)

Differences in Pathway Abundance Associated with 'Age_binary' (ANCOM Results) (ITS)

Heatmaps

These heatmaps display the relative abundance of predicted functional pathways across different samples, based on PICRUSt2-inferred metagenomic profiles from 16S rRNA gene data. Generated using QIIME2, each row represents a metabolic pathway (from MetaCyc), and each column corresponds to a sample group. The colour intensity indicates the predicted abundance of each functional pathway in the respective group.

You can click on each image to view a larger version in a new tab.

---

Fungi (ITS)

Results Summary

Fungal functional diversity was primarily influenced by pH and establishment method. Lower functional diversity was observed in alkaline soils. Guild analysis showed dominance of undefined saprotrophs, with dung-associated and pathogenic guilds more prevalent in managed or disturbed sites.

Functional Alpha Diversity

For fungal communities (ITS), functional diversity was assessed using metrics derived from predicted ecological functions (e.g., trophic modes, symbiosis potential).

The metrics below—Shannon diversity and Evenness—summarise the variability of ecological roles across different experimental conditions. Use the dropdown menu to explore patterns by variable.

Results Summary

Establishment method had no significant impact on fungal guild diversity, whether measured by Shannon index or evenness. In contrast, soil pH significantly influenced both metrics, with more acidic to neutral soils (below median pH) showing higher functional diversity than alkaline sites (Shannon: p-adjusted = 0.003; Evenness: p-adjusted = 0.007).

Neither site age nor management practices (e.g., grazing, ploughing, cutting) had a significant effect on functional diversity, although younger sites tended to have slightly lower diversity overall.

Select variable: Establishment Cutting Cattle Sheep Plough Age category pH category Age

Shannon Index Boxplots

Here is a link to the Shannon Index Boxplot for more flexibility on QIIME2: Shannon Index Boxplot - Functional (ITS)

Kruskal-Wallis p-value: 0.529

Kruskal-Wallis p-value: 0.797

Kruskal-Wallis p-value: 0.133

Kruskal-Wallis p-value: 0.592

Kruskal-Wallis p-value: 0.161

Kruskal-Wallis p-value: 0.439

Kruskal-Wallis p-value: 0.00299

Kruskal-Wallis p-value: 0.372

Evenness Index Boxplots

Here is a link to the Evenness Boxplot for more flexibility on QIIME2: Evenness Boxplot - Functional (ITS)

Kruskal-Wallis p-value: 0.226

Kruskal-Wallis p-value: 0.349

Kruskal-Wallis p-value: 0.106

Kruskal-Wallis p-value: 0.617

Kruskal-Wallis p-value: 0.331

Kruskal-Wallis p-value: 0.487

Kruskal-Wallis p-value: 0.00668

Kruskal-Wallis p-value: 0.206

Functional Beta Diversity

Results Summary

Analysis of fungal guild composition across the sites revealed significant effects of establishment method, soil pH, and site age, with patterns consistent across both Bray-Curtis and Jaccard distance metrics.

PCoA plots showed partial clustering by establishment method, with natural regeneration (NR) sites forming a tighter group, while green hay (GH) and seed mix (SM) plots were more dispersed. Despite some overlap, NR sites displayed more consistent guild structure. Similar trends were observed for both Bray-Curtis and Jaccard indices.

Clustering by pH was more evident in Jaccard-based PCoAs, where sites with slightly acidic to neutral pH values (below median) grouped more tightly than those with alkaline soils. Bray-Curtis plots revealed less pronounced patterns.

For site age, clustering was visible but overlapping, with two main groups composed of both younger (<13 years) and older (>13 years) sites, suggesting that while age influences guild composition, some core functional groups remain across time.

PERMANOVA results confirmed these observations:

• Establishment method: significant for both Bray-Curtis (p = 0.003) and Jaccard (p = 0.002), with all pairwise comparisons significant. The strongest differences were between NR and GH/SM sites.
• Soil pH category: significant for both metrics, but stronger with Jaccard (Bray-Curtis p-adjusted = 0.003; Jaccard p-adjusted = 0.001).
• Age category: also significant (Bray-Curtis p-adjusted = 0.009; Jaccard p-adjusted = 0.04), supporting the role of successional dynamics in shaping functional guild structure.

These findings indicate that functional beta diversity in fungal communities responds sensitively to restoration strategy, edaphic conditions, and ecological time, with Jaccard providing slightly better resolution for detecting presence/absence-driven changes in community function.

Bray-Curtis

Emperor Plot

Here is a link to the Bray-Curtis Emperor Plot for more flexibility on QIIME2: Bray-Curtis Emperor Plot (16S)

Bray-Curtis Emperor Plot (Fungi - ITS, Functional)

PERMANOVA

You can click on the images below to access the full QIIME2 report.

Select variable: Establishment pH category Age category

View full QIIME2 results (Bray–Curtis – Establishment)

Figure: PERMANOVA Bray-Curtis for “Establishment” (ITS)

View full QIIME2 results (Bray–Curtis – pH)

Figure: PERMANOVA Bray-Curtis for “pH” (ITS)

View full QIIME2 results (Bray–Curtis – Age)

Figure: PERMANOVA Bray-Curtis for “Age” (ITS)

Jaccard

Emperor Plot

Here is a link to the Jaccard Emperor Plot for more flexibility on QIIME2: Bray-Curtis Emperor Plot (16S)

Jaccard Emperor Plot (Fungi - ITS, Functional)

PERMANOVA

You can click on the images below to access the full QIIME2 report.

Select variable: Establishment pH category Age category

View full QIIME2 results (Jaccard – Establishment)

Figure: PERMANOVA Jaccard for “Establishment” (ITS)

View full QIIME2 results (Jaccard – pH)

Figure: PERMANOVA Jaccard for “pH” (ITS)

View full QIIME2 results (Jaccard – Age)

Figure: PERMANOVA Jaccard for “Age” (ITS)

Fungal Functional Guilds (FUNGuild)

In microbial ecology, a guild refers to a group of organisms that fulfil similar ecological roles, regardless of their taxonomic identity. Understanding functional guilds allows researchers to move beyond taxonomic profiles and assess the ecological roles that microbial communities may play in an environment.

To investigate the ecological roles of fungal communities, we used FUNGuild, a tool that assigns fungi to functional guilds based on curated databases and literature. These guilds represent ecological strategies such as:

• Saprotrophs: decomposers of organic matter
  
• Mycorrhizal fungi: symbionts associated with plant roots
  
• Pathogens: organisms that cause disease in plants or animals
• Ectomycorrhizal: symbionts that assist plant roots by enhancing nutrient and water uptake, especially nitrogen and phosphorus, while contributing to carbon cycling and soil nutrient mobilisation
• Arbuscular Mycorrhizal: symbionts that contribute to nutrient uptake
• Endophyte: microorganisms that promote the growth and development of the plants
• Lichenized: fungi that form symbiotic associations with algae or cyanobacteria, creating lichens that can survive in harsh environments by combining structural support and photosynthetic ability
• Parasites: fungi that live on or inside a host organism, extracting nutrients and often causing harm
• Symbiotrophs: fungi that engage in mutually beneficial relationships with host organisms, typically exchanging nutrients for resources like carbon or protection

This functional classification provides valuable insights into what fungi are likely doing in the ecosystem, beyond simply who they are.

This section explores the functional roles of fungi within each site, based on guild-level annotations provided by FUNGuild. Fungal guilds reflect ecological functions such as saprotrophy, symbiosis (e.g., mycorrhizal fungi), or pathogenicity. This approach provides insight into how fungal communities may contribute to ecosystem processes, complementing traditional taxonomic analyses.

Results Summary

The analysis of fungal guild abundance across the sites revealed high variability in fungal ecological roles. A consistent observation was the substantial proportion of unassigned fungi, limiting interpretation in some cases. Among the identified guilds, the most dominant were:

• Undefined saprotrophs
• Plant pathogens
• Dung saprotrophs

In select sites, animal parasites were also present at noteworthy levels.

Across site age categories, distinct patterns emerged. In older sites (above median age), the fungal community was dominated by undefined saprotrophs, comprising nearly 50% of the total abundance. Plant pathogens and dung saprotrophs were also significantly represented. In younger sites, while undefined saprotrophs remained abundant, plant pathogens and dung saprotrophs were present in relatively higher proportions. In addition, animal pathogens and wood saprotrophs were detected in younger sites.

When comparing establishment methods, the guild composition also varied:

• Green hay (GH) sites showed high levels of undefined saprotrophs, plant pathogens, and dung saprotrophs.
• Natural regeneration (NR) sites had the highest percentage of undefined saprotrophs, but lower levels of plant pathogens and dung saprotrophs.

Regarding soil pH, sites with above-median pH displayed high abundances of undefined saprotrophs, followed by plant pathogens and dung saprotrophs. In contrast, below-median pH sites showed a similar dominance of undefined saprotrophs, but with an increased proportion of animal parasites.

These patterns suggest that fungal functional composition is responsive to restoration strategy, soil chemistry, and successional stage.

Top 20 Most Abundant Guilds

The plot below highlights the top 20 most abundant fungal guilds identified using FUNGuild. To avoid clutter, the guild names are hidden on the y-axis; however, users can hover over each bar to reveal the full name, enabling interactive and detailed exploration of fungal functional diversity.

Top 20 functional guilds

Guild Abundance Across Sites

The figure below shows the total abundance of fungal ASVs across sites, aggregated by functional guild. This provides an overview of how guild-level composition varies between locations, which may reflect differences in land use, soil conditions, or restoration histories.

Fungal communities were dominated by Undefined Saprotrophs, Plant Pathogens, and Dung Saprotrophs. A high proportion of Unassigned fungi highlights limits in current taxonomic resolution.

Fungal Guild Abundance by Site

Guild Abundance by Variable

Establishment

Natural regeneration had the highest proportion of Undefined Saprotrophs, while Green Hay/Bush and Seed Mix had more Plant Pathogens and Dung Saprotrophs.

pH

Alkaline sites showed dominance of Undefined Saprotrophs; acidic sites had more Animal Parasites.

Age

Older sites had more Undefined Saprotrophs; younger sites showed higher proportions of Plant Pathogens and Dung Saprotrophs.

Select variable: Establishment Age pH

Fungal Guild Relative Abundance by pH_category

Fungal Guild Relative Abundance by Age_category

Fungal Guild Relative Abundance by Establishment

---

Bacterial and Fungal Comparison

Results Summary
Correlations between bacterial and fungal taxa and functions revealed potential ecological links, particularly between saprotrophic/pathogenic fungi and sugar-degrading or nitrogen-associated bacterial pathways. Most associations were site-specific and stronger in NR plots, with nutrient cycling emerging as a shared functional axis.

Taxonomic level

After filtering for the most abundant taxa, seven significant correlations were detected at the family level, with Pearson correlation coefficients ranging from -0.61 to 0.80. Among these, six were positive and one was negative.

The main bacterial families involved were:

• Bradyrhizobiaceae
• Thermomonosporaceae
• Rhodospirillaceae
• Chthoniobacteraceae
• Rhodobacteraceae

The corresponding fungal families included:

• Trimorphomycetaceae
• Clavicipitaceae
• Tricholomataceae
• Piskurozymaceae
• Pseudeurotiaceae

At the class level, dominant bacterial groups were Alphaproteobacteria, Actinobacteria, and Spartobacteria. Fungal communities were mainly composed of Tremellomycetes, Sordariomycetes, Agaricomycetes, and Leotiomycetes.

Below is an interactive visualisation of significant correlations between abundant bacterial and fungal families:

Click to view the detailed correlation table

Significant correlations between bacterial and fungal taxa

Bacteria	Fungi	r	p_adj
Bradyrhizobiaceae	Trimorphomycetaceae	0.80	0.0000000
Bradyrhizobiaceae	Clavicipitaceae	0.79	0.0000000
Thermomonosporaceae	Trimorphomycetaceae	0.73	0.0000007
Rhodospirillaceae	Tricholomataceae	0.72	0.0000022
Thermomonosporaceae	Piskurozymaceae	0.72	0.0000028
Chthoniobacteraceae	Pseudeurotiaceae	0.72	0.0000039
Rhodobacteraceae	Clavicipitaceae	-0.61	0.0118590

---

Functional level

A total of 42 significant correlations were identified between bacterial pathways (from the MetaCyc database) and fungal guilds (from FUNGuild). Among these, 40 correlations were positive, and 2 were negative, with Pearson correlation values ranging from -0.75 to 0.98.

While most correlations involved rare pathways or guilds (appearing at only one or two sites or with low overall counts), 10 correlations stood out due to their association with more abundant functions.

The dominant fungal guilds among these included:

• Animal pathogen / dung saprotrophs
• Endophytic / ericoid saprotrophs
• Fungal and insect parasites

The correlated bacterial pathways primarily involved sugar and protein degradation, such as:

• Methylglyoxal degradation
• L-arabinose degradation
• Glucose, sucrose, and L-tryptophan degradation
• Purine degradation

Below is an interactive visualisation of significant correlations between abundant bacterial pathways and fungal guilds:

Click to view the detailed table for the 10 strongest correlations

Significant correlations between bacterial pathways and fungal guilds

Guild	Pathway	r	p_ajd
Animal pathogen / dung saprotroph	Methylglyoxal degradation	0.94	0.0e+00
Animal pathogen / dung saprotroph	L-arabinose degradation IV	0.85	0.0e+00
Animal pathogen / dung saprotroph	Glucose and glucose-1-phosphate degradation	0.80	0.0e+00
Animal pathogen / dung saprotroph	Sucrose degradation IV (sucrose phosphorylase)	0.78	0.0e+00
Endophyte–Ericoid Mycorrhizal–Undefined Saprotroph	Superpathway of N-acetylneuraminate, N-acetylglucosamine, and N-acetylmannosamine degradation to β-D-fructofuranose 6-phosphate	0.74	1.0e-07
Animal pathogen / dung saprotroph	L-tryptophan degradation to 2-amino-3-carboxymuconate semialdehyde	0.72	8.0e-07
Animal pathogen / dung saprotroph	2-methylcitrate cycle II	0.71	1.4e-06
Endophyte–Ericoid Mycorrhizal–Undefined Saprotroph	Superpathway of N-acetylneuraminate degradation	0.71	2.4e-06
Dung Saprotroph–Ectomycorrhizal–Litter Saprotroph	Purine nucleobases degradation I (anaerobic)	-0.71	2.6e-06
Endophyte–Epiphyte–Fungal Parasite–Insect Parasite	Nylon-6 oligomer degradation	0.71	3.0e-06

Discussion

Bacterial

Alpha and beta diversity

Microbial alpha diversity was significantly influenced by restoration establishment methods, with GH and SM plots showing higher phylogenetic and taxonomic diversity than NR, as indicated by Faith’s phylogenetic diversity, Pielou’s evenness, and Shannon entropy. These results suggest assisted restoration approaches promote more diverse microbial communities, likely due to increased plant richness, root exudate variation, and organic inputs associated with sown vegetation (Van der Putten et al., 2013; Bardgett & van der Wal, 2014). In contrast, site pH, management type, and restoration age had no significant effect on alpha diversity, implying that richness and evenness remain relatively stable once communities are re-established, potentially buffered by functional redundancy among microbial taxa (Louca et al., 2018).

In terms of beta diversity, all four tested environmental factors—restoration method, site age, pH, and grazing—significantly influenced microbial community structure. Restoration method was the strongest driver, with clear differences in both community membership (Unweighted UniFrac, p = 0.001) and relative abundance (Bray-Curtis, p = 0.001), supporting the idea that assisted establishment strategies select for distinct microbial assemblages. These differences likely reflect the influence of plant-driven processes and disturbance regimes on microbial recruitment and filtering. Site age had a more moderate effect, primarily altering abundance patterns rather than community membership (Bray-Curtis p = 0.037; UniFrac p = 0.063), consistent with successional theory, which suggests that microbial communities become functionally more specialised over time rather than undergoing wholesale taxonomic turnover (Fierer et al., 2010; Nemergut et al., 2013).

Soil pH emerged as a strong environmental filter, affecting both composition and phylogenetic diversity (Bray-Curtis p = 0.001; UniFrac p = 0.002). Even moderate changes in soil pH led to noticeable differences in microbial communities. This is consistent with previous studies showing that pH shapes microbial composition by influencing nutrient availability, enzyme activity, and microbial stress responses (Lauber et al., 2009; Rousk et al., 2010). Grazing also significantly impacted community composition (PERMANOVA p = 0.001), likely via soil disturbance, altered plant cover, and localized nutrient enrichment from faeces and urine (Bardgett & Wardle, 2010; Patra et al., 2005). These findings highlight the combined importance of biotic management and edaphic conditions in shaping microbial diversity, with restoration strategy exerting strong initial filtering and environmental variables driving longer-term compositional dynamics.

Functional diversity

Microbial functional beta diversity varied significantly across environmental gradients, indicating community function is responsive to restoration strategy and site conditions. Among all tested variables, restoration method exerted the strongest influence on functional composition (PERMANOVA p = 0.001), with clear differences in predicted pathway profiles between NR, SM, and GH plots. These differences likely reflect changes in rhizosphere conditions, litter chemistry, and substrate availability introduced by different plant communities (Eisenhauer et al., 2017; Zhalnina et al., 2018). Functional divergence under assisted restoration suggests that microbial communities assemble not only taxonomically but also functionally in response to the vegetation structure and resource environment shaped by establishment type (Lange et al., 2015).

Soil pH also significantly shaped microbial function (PERMANOVA p = 0.012), reinforcing its role as a major ecological filter. Functional differences between high- and low-pH sites may result from pH-driven shifts in microbial physiology, including changes in nutrient acquisition strategies and metabolic activity (Lauber et al., 2009; Rousk et al., 2010). For instance, alkaline soils favored pathways linked to secondary metabolism, such as pentalenolactone biosynthesis, associated with antimicrobial production in Actinobacteria (Tetzlaff et al., 2006).

Site age had a more modest yet significant effect on functional beta diversity (PERMANOVA p = 0.035), with older sites exhibiting functionally distinct communities. Younger sites showed enrichment in core biosynthetic pathways such as heme and methionine biosynthesis, consistent with early successional stages where microbial populations prioritise metabolic versatility and growth (Choby & Skaar, 2016; Mohany et al., 2021). Over time, these communities are likely to transition toward greater functional specialisation and redundancy, a pattern predicted by successional models of microbial assembly (Fierer et al., 2010; Nemergut et al., 2013; Delgado-Baquerizo et al., 2020).

Grazing also influenced microbial functional composition (PERMANOVA p = 0.035), with grazed soils enriched in pathways related to nitrogen turnover and organic matter degradation, including glutaryl-CoA and glutamate degradation (Wang et al., 2020; Tang et al., 2020). These patterns reflect the combined effects of physical disturbance, nutrient enrichment, and vegetation shifts under grazing pressure (Patra et al., 2005; Bardgett & Wardle, 2010).

Differential functional pathways

The differential abundance of specific microbial pathways across environmental gradients highlights how microbial communities adapt functionally to changes in soil conditions and management. In alkaline soils, two secondary metabolite biosynthesis pathways—pentalenolactone (PWY-6915) and neopentalenoketolactone/pentalenate (PWY-6919)—were enriched, suggesting that higher pH environments may favor Actinobacteria capable of producing antimicrobial compounds (Tetzlaff et al., 2006; Lauber et al., 2009) potentially conferring a competitive advantage in resource-limited or less acidic environments, indicating microbial adaptation to changes in soil chemistry. (Rousk et al., 2010).

Site age also influenced key biosynthetic pathways, with younger sites enriched in functions related to heme and methionine biosynthesis which support oxidative metabolism and protein synthesis. This indicates that early-stage microbial communities prioritise metabolic generalism and resilience during colonisation (Choby & Skaar, 2016; Mohany et al., 2021). This supports successional theory, where functional activity in early communities is geared toward rapid establishment and broad resource use, gradually shifting toward specialisation over time (Fierer et al., 2010; Nemergut et al., 2013).

The restoration method further shaped microbial functional potential. SM and GH plots showed higher relative abundance of peptidoglycan biosynthesis (PWY-5265) and thiamine diphosphate biosynthesis (PWY-6892), pathways associated with cell wall integrity and cofactor production, respectively (Heijenoort, 2001; Begley et al., 1999). These functions may reflect higher microbial growth rates and metabolic activity supported by richer vegetation inputs in assisted restoration systems (Eisenhauer et al., 2017; Lange et al., 2015).

These pathway-level shifts demonstrate that soil microbial communities respond to environmental variation with targeted functional adaptations. The enrichment of specialized metabolic traits under different pH, age, establishment, and grazing conditions reflects the dynamic role of microbes in mediating ecosystem processes under changing land management regimes.

Core metabolic functions maintained across sites

Despite clear differences in microbial community structure and specialised metabolic pathways across environmental gradients, a conserved set of core functions was consistently maintained. The top 10 most abundant MetaCyc pathways were shared across all samples regardless of site age, soil pH, or restoration method, highlighting the persistence of key biosynthetic and energy-generating processes. These included pathways involved in branched-chain amino acid biosynthesis, pyruvate fermentation and aerobic respiration. Their ubiquity suggests microbial communities retain a stable functional core essential for survival and basic ecosystem functioning.

This pattern supports the concept of functional redundancy, where different microbial taxa may fulfill similar roles, ensuring continuity of core functions even as community composition or secondary pathways shift (Louca et al., 2018). While environmental factors drive differentiation in specific metabolic traits, the persistence of conserved pathways implies a buffering capacity within soil microbial systems that may contribute to ecosystem resilience. While microbial communities respond flexibly to environmental and management changes, their foundational metabolic roles remain remarkably robust.

Taxonomical composition

Actinobacteria were abundant across all soils, aiding organic matter decomposition (Bao et al., 2021). The dominance of copiotrophic Proteobacteria in potentially nutrient-rich SM and GH soils, could reflect their preference for environments with abundant nutrients (Spain, Krumholz, & Elshahed, 2009). Acidobacteria, linked to low-nutrient, acidic soils, were consistently present (~5–16%) (Jones et al., 2009), although the lower pH sites in this study were only slightly acidic.

Chloroflexi and Verrucomicrobia were less abundant overall, though Spartobacteria (in Verrucomicrobia) increased notably in NR soils. Verrucomicrobia is abundant in grassland soils and has a slow growth rate making it adaptable for exploiting sparse resources (Bergmann et al, 2011) in NR sites.

Above-median pH soils were dominated by Actinobacteria and Proteobacteria (~40–50%), while below-median pH soils showed higher Firmicutes (~25–35%), suggesting acid tolerance. Minor phyla were slightly more frequent in low pH soils, indicating greater microbial diversity. These trends align with models where copiotrophs thrive in higher pH, and oligotrophic, acid-tolerant taxa persist in acidic environments (Fierer, Bradford, & Jackson, 2007). Table 5 shows a breakdown of the ecological roles performed by different taxa.

Table 5: Taxonomical classification and ecological roles for the major bacterial taxa found in soils across the sites

Phylum	Class	Order	Role
Actinobacteria	Actinobacteria	Actinomycetales	Decomposers of tough organic material; drought-tolerant
Firmicutes	Bacilli	Bacillales	Stress-tolerant, spore-forming bacteria (e.g., Clostridia)
Proteobacteria	Alphaproteobacteria, Gammaproteobacteria	Rhizobiales, Pseudomonadales	Fast growth, nutrient-responsive bacteria
Verrucomicrobia	Verrucomicrobiae	Bacillales	Soil-associated oligotrophs, survive in low nutrient conditions
Acidobacteria	Acidobacteria	Bacillales	Abundant in acidic, low-nutrient soils; important in organic matter decomposition and soil stabilization
Chloroflexi	Anaerolineae	Anaerolineales	Slow degraders in moist microenvironments
Planctomycetes	Planctomycetacia	Planctomycetales	Biofilm formers, complex carbon cycle contributors
Bacteroidetes	Bacteroidia	Bacteroidales	Break down complex carbohydrates
Nitrospirae	Nitrospira	Nitrospirales	Nitrogen cycle bacteria (nitrification)

The significantly higher abundance of Conexibacteraceae in slightly acidic to neutral soils has been found in other studies and is attributed to a potential ecological preference for lower pH environments (Lauber et al., 2009; Fierer & Jackson, 2006) although there are few studies on their role in grassland systems. The bacterial family Intrasporangiaceae alongside Blastococus, showed preferences for sites seeded with GH and SM. Both families belong to Actinobacteria, a major decomposer. Their presence on sites with active restoration strategies suggests they respond positively to enhanced plant inputs and nutrient accessibility associated with planted vegetation (Jangid et al., 2011). This further supports the suggestion that assisted restoration methods promote more diverse microbial communities (Van der Putten et al., 2013; Bardgett & van der Wal, 2014).

Fungi

Alpha and beta diversity

The results for fungi partially support the hypothesis that age, establishment method, management type and pH have a significant effect on alpha and beta diversity in fungal communities. Of all the drivers, pH was the most significant for both alpha and beta diversity, followed by age. Establishment method was also a strong driver for the differences between sites.

The impact of pH on fungal diversity is strongly supported by other studies, although the literature is contradictory. Ma et al (2023) found a negative correlation between soil pH and Shannon diversity when comparing grasslands with soils ranging from pHs of 6 to 7, although the study focused on grasslands in China, limiting direct comparisons with this study. However, Seaton et al (2022) found a positive correlation when comparing sites with pH values ranging from 6 to 8. In this study sites were split between slightly acidic to neutral (pH 6.6-7.9) and alkaline (pH 7.91-8.63) with alkaline soils showing lower alpha diversity, contradicting the findings of Seaton et al (2022). Reduced fungal diversity in soils with high pH have been attributed to the reduced nutrient use efficiency of fungi in these soils leading to a reduction in fungal diversity (Ma et al, 2023). However, it is difficult to draw this conclusion in this study.

For age, this study found that younger sites had lower alpha diversity than middle-aged to older regeneration sites and that age was a major driver for fungal community differences. Fungal communities take longer to recover from ecosystem disturbance (Gao et al, 2021) therefore may require longer-term restoration efforts, demonstrated by sites in the more diverse, above median age category in this study ranging in age from 16 to 100 years.

Differential Abundance Analysis

Establishment method had a strong effect on differences in fungal composition between sites supported by the beta diversity PERMANOVA results. Differential abundance analysis showed Metarhizium was almost absent from GH and SM sites, but more frequent in NR plots. Metarhizium is an entomopathogen that infects insects. Its presence in less disturbed plots NR might be linked to more stable host populations and fewer changes in soil structure (Zimmermann, 2007). This suggests a preference for undisturbed environments or stable host availability. Despite its known resilience to harsh soil conditions (Vänninen et al., 2000), its absence in disturbed sites points to sensitivity to competitive or structural changes in soil. On the other hand, Gibellulopsis was barely detected in NR plots but found in both GH and SM sites. Although no ecological studies specifically addressing the genus Gibellulopsis were found, Gibellulopsis nigrescens (syn. Verticillium nigrescens) has been reported as a weak plant pathogen involved in Verticillium wilt on peppermint in Germany (Alisaac & Götz, 2022). While this is a highly specific case, it illustrates how certain fungal taxa may respond to plant-related or environmental conditions.

Sheep grazing was also shown to be a significant factor affecting differences in soil fungal communities and was associated with higher abundance of Helotiales_fam_Incertae_sedis and Preussia. The observed increase in Helotiales_fam_Incertae_sedis under sheep grazing may reflect a response to changes in plant-derived inputs (e.g., litter or root exudates) induced by herbivory. Although the ecological roles of many Helotiales taxa remain unresolved—with approximately 70% of species lacking clear lifestyle or ecological data (Bruyant, Moënne-Loccoz, & Almario, 2024)—the order includes a broad diversity of functional types such as saprotrophs, endophytes, and mycorrhizal symbionts (Rissi, Ijaz, & Baschien, 2024). Sheep contribute to soil dynamics via dung and urine deposition, which can alter soil chemistry and nutrient availability. Although this effect is mainly documented in the context of greenhouse gas emissions, it implies a modification of soil organic inputs (Jiang et al., 2014). Some Preussia species have been isolated from dung (Kruys et al., 2006), suggesting that at least some members of the genus may benefit from such inputs.

Ploughing also affected some fungal taxa, notably Saitozyma, a saprotrophic yeast more commonly found in ploughed plots. Genomic studies have shown that Saitozyma podzolica (previously Cryptococcus podzolicus) possesses the metabolic capacity to break down plant-derived carbon sources such as glucose and xylose and convert them into lipids and gluconic acid (Aliyu et al., 2021). These traits suggest that Saitozyma may benefit from the increased availability of labile substrates and altered soil conditions caused by ploughing, enabling it to rapidly colonise and exploit disturbed soils. Xenopyrenochaetopsis and Gloniaceae (gen. Incertae sedis) were also more common in plots disturbed by ploughing, suggesting they may behave as saprotrophs or endophytes capable of exploiting decomposing material or altered microhabitats. However, due to the limited documentation available on their functional roles or ecological preferences, any such interpretation remains speculative.

Fungal composition also shifted with site age. Paraphaeosphaeria was only found in older sites, though in low abundance. This genus is known as an endophyte, sometimes linked to biocontrol functions (Baroncelli et al., 2020). In contrast, Lasiosphaeriaceae_gen_Incertae_sedis was more abundant in younger sites. Members of this fungal family are frequently described as saprotrophs, including species associated with dung or decaying vegetation (Louw et al., 2022), which may allow them to benefit from the high organic input and disturbance typical of early successional systems.

Finally, we observed that Metapochonia was more frequently detected in soils with lower pH values ( <7.94). Although this pattern suggests a possible ecological preference, there is currently no clear documentation on the ecological role or functional traits of this genus in acidic environments. Further studies are needed to clarify the ecological function of Metapochonia in soil systems.

Differential abundance of functional guilds

The results on fungal functional guilds support the hypothesis that site age, establishment method, and soil pH influence the ecological structure of fungal communities, although taxa identification is limited by the lower resolution of amplicon sequencing. Among these factors, establishment method appeared to exert the strongest influence, particularly distinguishing NR sites from those established via GH or SM.

These patterns likely reflect the influence of disturbance intensity and grazing regime. NR sites, which experience reduced intervention, exhibited the lowest relative abundance of plant pathogens and dung saprotrophs. In contrast, sown plots showed markedly higher proportions of these guilds, potentially due to increased herbivore activity, elevated plant biomass, or initial organic matter inputs. This supports previous findings that fungal functional composition responds sensitively to land management and vegetation structure (Hannula et al., 2017).

The influence of soil pH was subtler but notable. Although overall guild diversity remained stable, animal parasites were more abundant in acidic soils, suggesting pH-dependent shifts in host-parasite dynamics or soil biodiversity patterns. This observation aligns with evidence that soil chemistry—including pH and salinity—can shape fungal guild composition, sometimes enhancing the abundance of specific groups (Chandran et al., 2023).

We observed a relative influence of site age on fungal community composition in our grassland system, similar patterns have been reported in forest restoration contexts. For instance, in subalpine Picea asperata plantations, stand age was identified as the main driver of variation in fungal community composition and functional guilds (Fang et al., 2023). While these findings support the idea that successional time can influence soil fungal communities, they should be interpreted with caution given the ecological differences between forest systems and grasslands. The extent to which such patterns generalise across ecosystems likely depends on vegetation type, soil properties, and dominant fungal functional groups.

A recurrent and striking pattern across all site types was the dominance of saprotrophic fungi, particularly undefined saprotrophs. This suggests a strong functional role in decomposition and nutrient cycling, typically associated with later successional stages (Peay et al., 2016). However, the large proportion of unassigned fungi also underscores the limitations of current fungal functional classification systems (Nilsson et al., 2020), potentially obscuring a portion of the ecological signal.

Bacterial and fungal interactions

This study illustrates that bacterial and fungal communities are affected by different drivers. Diversity within bacterial communities is largely driven by establishment methods while fungal diversity is significantly affected by soil pH and age. However, drivers for the differences in community composition (beta) between sites were broadly similar for bacterial and fungal communities, with establishment, pH, age and to a more limited extent, management the most important factors. A look at differentially abundant taxa between the two kingdoms shows taxonomical composition patterns. This includes preferences for GH and SM sites, that tend to have more diverse plant communities, by copiotrophic bacteria such as Proteobacteria and fungi Gibellulopsis. At a pathway and guild level core metabolic pathways for bacteria and core decomposition functions from saptrophs also seem to dominate. Making comparisons between bacterial and fungal communities is however made difficult by the shortage of documentation associated with fungal taxonomy and function.

Both fungal and bacterial communities are important for multi-functionality in ecosystems, supporting functional diversity as well as functional redundancy, where an ecosystem contains multiple species that provide the same function. Functional redundancy acts as a ‘future proofing’ for ecosystems, as it protects against the loss of taxa and ensures functioning is maintained (Wagg et al, 2019). Relationships between fungi and bacteria can be complementary or supportive (Wagg et al, 2019; Kohlmeier et al, 2005).

In this study, the most significant taxa correlations were between saprotrophic or parasitic/pathogenic fungi and heterotrophic or organotrophic bacteria. The dominance of these groups is often found in grasslands (Labouyrie et al, 2023). For functional correlations, most of the significant relationships were between animal pathogens and saprotrops and microbial metabolic pathways associated with the breakdown of sugars and other macromolecules.

Placing specific correlations in the context of existing literature proved challenging, chiefly due to the limited nature of many studies. Although there are multiple studies considering the effects of fungal-bacterial interactions on ecosystem function, literature is still sparse with those studies that are available treating bacterial and fungal diversity separately (Labouyrie et al, 2023; Lepinay et al, 2024), drawing general conclusions (Gao et al, 2021; de Menezes et al., 2015), or investigating interactions in non-grassland ecosystems (Gomez-Brandon et al, 2020).

Taxa correlations

In this study, there was some evidence for fungal-bacterial interactions at a taxa level. For instance, Bradyrhizobiaceae, an oligotrophic family of nitrogen-fixing, phosphate-mobilising bacteria was correlated with the abundance of saprotrophic yeasts (genus: Saitozyma) from the Trimorphomycetaceae family, and genus Keithomyces, an insect parasite from the pathogenic fungal family Clavicipitaceae. The relative abundance of saprotrophic fungi, including yeasts, responds positively to increased bio-available nitrogen (Gomez-Brandon et al, 2020; Simonin et al, 2017) as do fungal pathogens (Geng et al, 2023) so it is possible the increased prescence of these fungi on Bradyrhizobiaceae-containing sites could be related to increased nitrogen and phosphorous availability. However, soil nitrogen levels at these sites would need to be assessed to confirm this. An alternative explanation is that pH levels were a driving factor as all three taxa appeared on the sites with the lowest pH values (6.6-6.7). Bradyrhizobiaceae in particular favours acidic soils (Sawada et al, 2023).

Many of the taxa correlations appeared to be driven by site preferences. For example, Rhodospirillaceae, a family of photoheterotrophic bacteria appeared to be correlated with Dermaloma, as both taxa were most abundant on one site, Knepp Pond Field. Rhodospirillaceae prefers anoxic conditions while saprotroph Dermaloma, prefers unimproved grassland (O’Reilly and Parker, n.d.). It is possible therefore the positive correlation is related to the NR site providing the most favourable conditions for both taxa to thrive but for different reasons.

Functional correlations

The main functional correlations were related to nutrient cycling, with fungi and bacteria carrying out complementary processes or recycling different macromolecules.

Six of the significant pathways involved dung saprotrophs, associated with genera from the Chaetomiaceae family, mainly Thermothielavioides. Thermothielavioides produces α-ʟ-arabinofuranosidase, an enzyme that breaks down plant polysaccharides to release ʟ-arabinose (Camargo et al, 2020) and the bacterial ʟ-arabinose degradation pathway. Other sugar degradation pathways significantly positively correlated with Thermothielavioides were the glucose and glucose-1-phosphate degradation pathway (Caspi et al, 2020), and sucrose degradation IV (sucrose phosphorylase), associated with the bacterial order Actinomycetales (Caspi et al, 2020). Pathways involving the breakdown of tryptophan and the degradation of proprionate (2-methylcitrate cycle II), a by-product of fermentation found in ruminants (Caspi et al, 2020), were also positively associated with these fungal taxa. These correlations could suggest fungi and bacteria can take on complementary functions by breaking down different plant polysaccharides, different macromolecules or bacteria degrading the byproducts of fungal processes. Support for complementarity between the kingdoms has been found by Wagg et al (2019) in experimental grassland systems with these complementary functions providing different limiting nutrients to plants and increasing plant productivity.

The purine nucleobases degradation I (anaerobic) pathway was associated with the saprotrophic and ectomycorrhizal guilds related to Pezizaceae, a fungal family that includes genera found in dung, rotting wood or litter. The purine degradation pathway releases carbon and nitrogen in oxygen poor environments (Caspi et al, 2020), so this may be an example of bacteria taking advantage of the accessibility of macronutrients released by fungal nutrient cycling.

Most of the correlations were associated with sites that use NR as an establishment method. Fungal-bacterial associations have been shown to increase in complexity along successional stages in grasslands (Gao et al., 2021) so this is a potential avenue for research.

Limitations

Carrying out pairwise comparisons between fungi and bacteria resulted in thousands of correlations. Although correlations were filtered for abundance it may have resulted in potentially filtering out meaningful correlations while including correlations which, upon analysis, could not be clearly understood, even when taking the current literature into account. Additionally, some of the FUNGuild guild assignments included multiple, non-specific guilds for the same assignment, making it confusing for analysis. Methods other than pairwise correlations such as generalised linear models (Wagg et al, 2019) could be investigated.

This study also considered each factor in isolation, potentially missing the effects of interactions between establishment method and age. For future studies, more advanced statistical methods could be used to take age into account when assessing the effects of establishment method.

Conclusions

This study demonstrates how bioinformatics pipelines and analysis can be used to study grassland ecosystem dynamics and their impact on bacterial and fungal communities. Sequencing of the 16S and ITS amplicons have enabled the answering of research questions about the impact of restoration methods, soil physico-chemical properties, length of restoration and management practices on bacterial-fungal assemblies. Additionally, they can help ecological restoration projects such as RestREco establish best practices for restoring and building resilience in degraded grasslands. 

This study determined that establishment method, pH and age were the biggest drivers of differences in bacterial and fungal communities, but that within-site bacterial taxonomic and functional diversity was significantly affected by establishment method while soil pH and age were the major factors affecting fungal diversity. Active restoration methods such as the planting of seed mixes appear to provide the biggest benefit to bacterial communities instead of a natural regeneration approach. Soil pH and grazing impact community composition while age affects functional specializations. Fungal communities respond more positively to longer-term restoration efforts while higher soil pH can have a negative effect on fungal diversity.  

The dominance of specific taxa in bacterial communities such as decomposers Actinobacteria and Proteobacteria demonstrated bacterial preferences for soils with high nutrient availability while some less abundant phyla showed preferences for soils with lower pH values. For fungal communities the differential abundance of certain taxa, including saprotrophs and pathogenic fungi, in sites using different restoration methods and land management systems demonstrated the impact of plant diversity, nutrient inputs and soil disturbance on fungal assemblies. 

Core bacterial pathways and fungal functions for survival, growth, organic matter breakdown and nutrient cycling were dominant throughout. However, the impact of establishment method, pH, age and management on function was more multi-faceted. Establishment and management method affected functioning for both kingdoms but for different reasons. For bacteria, both variables were associated with nutrient inputs driving growth, metabolism and organic matter breakdown pathways, while for fungi, they were associated with decomposition and disturbance. Soil pH shaped functional communities for both kingdoms but introduced competition for resources in bacteria, demonstrated by the abundance of antimicrobial producing bacteria in higher pH sites. The role of age in bacterial functions was related to more ‘core’ pathways geared towards growth and resource use in younger sites but specialization and functional redundancy in older sites. The functional shifts caused by age on fungal guild composition was less clear.

Inter-kingdom analysis showed correlations between different taxa and functions were strongly associated with complementary or opportunistic relationships including the dominance of different pathways and functions associated with macromolecule breakdown or bacteria and fungi taking advantage of nutrients being made more available by members of the other kingdom. 

Overall, this study demonstrates that the impact of establishment method, pH and age on ecological restoration is complex, with bacterial and fungal communities responding differently to the three factors. For bacterial communities, assisted restoration methods provide the best chance of promoting diversity while for fungal communities ensuring pH levels are not too high and restoration projects are given enough time to allow communities to form are the key. Additionally, efforts to improve fungal taxa and functional classification methods should be stepped-up to increase understanding of fungal functions and how bacteria and fungi interact.